pH Measurements - What's Your Routine?

[couchsending]

Stirring is fine. Need to homogenize the sample. If your meter is very slow to get to a stable number you most likely need a new probe. A reading should stabilize rather quickly.

Highly suggest the Milwaukee 101 or 102 meters. Best bang for the buck.

If you’re fermenting in a vessel where taking samples is easy (stainless conical) checking pH during fermentation can also tell you a lot about the vitality of your yeast.

 

[chilitom]

I use a thin-walled disposable plastic cup. I withdraw ~50 ml into the cup, then swish it around in an ice bath. It takes maybe 30 seconds to bring it down to room temperature. I use one of the handheld pH meters and find that if I SLOWLY move it back and forth in the sample I get the most stable readings. Also, I calibrate it just before using it every time. I use the little buffer powder pillows from Hach that make up 50 ml of buffer solution, so I always have fresh buffer solutions for calibration. I have like 100 of these that will last years.

 

[day_trippr]

Do you check pH further along in the process?[...]

Nope, I have never tested anything post-fermentation...

Cheers!

 

[BrewDrinkRepeat]

I also agree 100% that by the time you take a pH reading, it's far too late to do anything about it in terms of the mash (although you could make adjustments intended to influence the flavor of the finished beer -- I don't, but I'm not saying it should never be done).

Use that data for the next batch of that recipe, and then RDWHAHB.

I don't always even bother to take a measurement, but when I do it's no earlier than the 30 minute mark, and more often than not I end up doing it right before I lift the grain basket. (I'm a no-sparge Foundry brewer these days.) Again, I'm using it as a reference measurement, not an actionable one.

I use Bru'n Water and a Hach Pocket Pro+, and like many I've found the estimated pH to be well within the "close enough" range for me, rarely varying from the meter by more than 0.1 or so. I design my water additions to hit a specific pH, but if I miss by a little I just don't worry about it.

If I was shooting for 5.4 and got 5.5 or 5.3, that doesn't bother me in the least, and the effect on the finished beer is negligible for the most part. I might make a compensation adjustment for next time, or I might not. And I certainly don't make that decision until I've fully evaluated the finished beer.

Here's my take on mash pH in general: it is (of course) an important metric in the brewing process, but as long as you are in the generally-accepted mash range (5.2 - 5.6, maybe 5.8 for very dark roasty malty beers) the exact pH is far less important to the finished beer than getting the flavor components of the water chemistry correct.

Missing your pH by a tenth or even two may be noticeable, but hardly catastrophic and the batch will certainly not be ruined in any way; get your water chemistry wrong and the finished beer could be significantly off the mark, especially for bitter and/or very malty styles.

 

[balrog]

I take a 50ml syringe sample at 30m stir time and set aside to measure when cool.
My biggest diff between predicted and measured are lighter SRM beers, biggest overall diff when doing recipe with 40-50% wheat.

 

[Deadalus]

I take a 50ml syringe sample at 30m stir time and set aside to measure when cool.
My biggest diff between predicted and measured are lighter SRM beers, biggest overall diff when doing recipe with 40-50% wheat.
View attachment 744815

Good data there balrog.

Which are the two calculators in the legend? What does the rest of the data (higher than 9 SRM) look like? MME is not particularly heterogenous...there are more observations in the 3-5 range than at 7.5-9 (just three brews), could just be missing more extreme residuals there. A confidence interval on the mean of the residuals (bias) would be informative. BF has some potential outliers around 3 SRM. Is it possible they are related in some fashion, same recipe for instance?

There's an accuracy to the pH meter used but there's also model accuracy of the calculator. They just spit out a value right but if they are built from real world observations there's a prediction accuracy too.